forked from mrq/DL-Art-School
6084915af8
Adds support for GD models, courtesy of some maths from openai.
Also:
- Fixes requirement for eval{} even when it isn't being used
- Adds support for denormalizing an imagenet norm
|
||
|---|---|---|
| .. | ||
| __init__.py | ||
| colors.py | ||
| convert_model.py | ||
| distributed_checkpont.py | ||
| fdpl_util.py | ||
| gpu_mem_track.py | ||
| kmeans.py | ||
| loss_accumulator.py | ||
| numeric_stability.py | ||
| options.py | ||
| util.py | ||
| weight_scheduler.py | ||